预后营养指数与需要营养支持的日本老年重度卒中后吞咽困难患者死亡率之间的关联：一项回顾性队列研究

文献信息

中文摘要

背景： 预后营养指数（PNI）由血清白蛋白和淋巴细胞计数计算得出，在卒中后吞咽困难（PSD）老年患者中研究仍不充分。我们探讨了PNI是否与日本老年PSD患者的全因死亡率相关，并将其预后性能与C反应蛋白/白蛋白比值（CAR）进行比较。
方法： 这项单中心回顾性队列研究纳入2014年1月至2017年1月期间132例需要PEG或TPN的日本老年重度PSD患者。采用多变量Cox回归分析和Kaplan-Meier曲线估计PNI和CAR与全因死亡率的关联。我们通过亚组和敏感性分析评估稳健性。为评价对1年死亡率的预后判别能力，我们使用时间依赖受试者工作特征（ROC）曲线，并使用综合判别改善（IDI）和净重分类改善（NRI）评估增量预测价值。
结果： 队列平均年龄为80.7岁（女性78例，男性54例）。在完全校正模型中，多变量Cox回归显示PNI每增加1分与全因死亡率降低7%相关（HR = 0.93，95% CI：0.90-0.97）。与最低PNI三分位组（T1）相比，最高三分位组（T3）显示风险降低75%（HR = 0.25；95% CI：0.10-0.60）。对于1年死亡率，PNI的时间依赖AUC在数值上高于CAR（0.771 vs. 0.712；p = 0.16）。将PNI加入临床基线模型可显著改善判别能力（IDI = 0.055，95% CI 0.010-0.104，p < 0.001；NRI = 0.471，95% CI 0.126-0.957），而加入CAR则没有（IDI = 0.015，p = 0.31）。
结论： 在日本老年重度卒中后吞咽困难且需要PEG或TPN的患者中，PNI与死亡率呈负相关，并在标准临床变量之外提供了增量预后信息。PNI是否可用于该人群的风险分层尚需前瞻性和外部验证。

英文摘要

BACKGROUND: The prognostic nutritional index (PNI), calculated from serum albumin and lymphocyte count, has remained insufficiently investigated in older patients with post-stroke dysphagia (PSD). We investigated whether PNI was associated with all-cause mortality in older Japanese patients with PSD and compared its prognostic performance with that of the C-reactive protein/albumin ratio (CAR).
METHODS: This single-center retrospective cohort study included 132 older Japanese patients with severe PSD requiring PEG or TPN between January 2014 and January 2017. Multivariable Cox regression analysis and Kaplan-Meier curves were used to estimate the associations of PNI and CAR with all-cause mortality. We assessed robustness through subgroup and sensitivity analyses. To evaluate the prognostic discrimination for 1-year mortality, we used time-dependent receiver operating characteristic (ROC) curves and assessed incremental predictive value using integrated discrimination improvement (IDI) and net reclassification improvement (NRI).
RESULTS: The mean age of the cohort was 80.7 years (78 women, 54 men). In fully adjusted models, multivariable Cox regression showed that each 1-point increment in PNI was associated with a 7% reduction in all-cause mortality (HR = 0.93, 95% CI: 0.90-0.97). Compared with the lowest PNI tertile (T1), the highest tertile (T3) showed a 75% risk reduction (HR = 0.25; 95% CI: 0.10-0.60). For 1-year mortality, the time-dependent AUC was numerically higher for PNI than for CAR (0.771 vs. 0.712; p = 0.16). Adding PNI to a clinical baseline model significantly improved discrimination (IDI = 0.055, 95% CI 0.010-0.104, p < 0.001; NRI = 0.471, 95% CI 0.126-0.957), whereas adding CAR did not (IDI = 0.015, p = 0.31).
CONCLUSION: PNI was inversely associated with mortality in older Japanese patients with severe post-stroke dysphagia requiring PEG or TPN and provided incremental prognostic information beyond standard clinical variables. Whether PNI may be useful for risk stratification in this population requires prospective and external validation.